PART I
TECHNICAL AND REVIEW PAPERS (Contd.)

ESTIMATION OF THE ECONOMIC BENEFITS OF FISHING: A REVIEW NOTE

M. J. Stabler

Amenity Waterways Study Unit, Department of Economics, University of Reading, Whiteknights, Reading, England

ABSTRACT

Most interest in the economics of recreation has focussed on the assessment of demand and the evaluation of benefits derived from specific sites for activities which are not traded through markets. Economic methods most appropriate to the evaluation of fisheries, and their management, are examined and discussed, with especial reference to method papers submitted for the Symposium. It is argued that the economic contribution to the management of fisheries has been too theoretical and that no clear agreement exists as to what is the value of fishery resources. Furthermore, differences in U.S. and European circumstances indicate that no general method can be devised. Contact and joint research between U.S. economists and their European counterparts is rare. Categorizing methods into aggregative or specific, it is suggested that, given the nature and state of fishery resources, two are worth developing and likely to prove operationally effective and comprehensible to practitioners: (1) systems analysis within which household production functions and site evaluation are included, (2) direct questioning of anglers to establish their evaluation of benefits. Partly because of compartmentalized research within disciplines, and partly because of data problems, it is concluded that no effective method for estimating fishery benefits has yet been developed. To assist this development, it is recommended that: (1) there should be better communication between economists in the U.S. and Europe, (2) more data should be collected on an aggregative and specific resource basis as part of the normal planning and management of fishery resources, (3) resources should be evaluated using more than one method concurrently, (4) a multi-disciplinary approach should be adopted.

RÉSUMÉ

La plupart de l'intérêt dans l'économie des loisirs a été centré sur l'évaluation de la demande et sur celle des bénéfices dérivés de domaines spécifiques d'activité, qui ne sont pas négociés à travers des marchés. Les méthodes économiques les plus appropriées pour l'évaluation des pêcheries et de leur gestion sont examinées et discutées, avec référence spéciale aux publications à caractère méthodologique, soumises pour le Symposium. Il est soutenu que la contribution des sciences économiques à la gestion des pêcheries a été trop théorique, et qu'il n'existe pas de clair agrément à savoir quel est la valeur des ressources de la pêche. De plus, des différences entre les circonstances européennes et aux États-Unis, indiquent qu'aucune méthode générale ne peut être formulée. Le contact et la recherche conjointe entre les économistes américains et leur confrères européens sont rares. En catégorisant les méthodes en “agrégatives” ou “spécifiques”, il est suggéré qu'étant donné la nature et l'état des ressources des pêches, deux d'entre elles valent la peine d'être développées, et sont promptes à se prouver operationellement effectives et compréhensibles par les parcticiens: (1) l'analyse des systèmes, dans laquelle sont inclues les fonctions de la production familiale et l'évaluation des sites; (2) le questionnement direct des pêcheurs à la lignes, afin d'établir leur évaluation des bénéfices. En partie à cause du cloisonnement de la recherche à l'intérieur des disciplines, et en partie à cause de problèmes de données, il est conclu qu'aucune méthode efficace pour l'estimation des bénéfices de la pêche n'a encore été développée. Afin d'aider à ce développement, il est recommendé: (1) qu'il y ait une meilleure communication entre les économistes aux É.-U. et en Europe; (2) que d'avantage de données soient accumulées sur la base des ressources “agrégatives” et “spécifiques”, en temps que part de la planification et gestion normales des ressources de la pêche; (3) que les ressources soient évaluées en utilisant plus d'une seule méthode concurremment; (4) qu'une approche multidisciplinaire soit adoptée.

INTRODUCTION

What might be termed the economics of recreation, as an aspect of applied economics, is little more than 20 years old; indeed it is only in the seventies that real progress was made in its development. Though researchers working in Europe have made a valuable contribution to the subject, both the concepts and methods of analysis have been dominated by work done in the U.S.

The principal areas of research interest have been:

1. estimation of the level and pattern of participation in recreation and sport;

2. the evaluation of land and water use, with the main emphasis being on rural areas and the sea;

3. the estimation of benefits and costs, both user and non user;

4. site or facility demand studies;

5. the management of recreation resources, e.g. optimal use of resources, resolution of conflicts, pricing policy and efficiency in providing resources.

Of these areas the overwhelming interest has been in assessing demand and evaluating the benefits derived from specific sites. This has occurred partly because of economists' concern with the basic issues on the best use of scarce resources, but it is also the result of the theoretical and practical difficulties encountered in developing more aggregative models and analysing the complex interrelationships between demand and supply.

Economists have devoted their attention in demand studies to identifying the recreational benefits and costs and to placing a monetary value on these. The problem is a difficult one where recreation is provided by the public sector at zero or nominal prices, as is the case with many fishery resources. However, even where a market exists the price of fishing may not reflect the cost of provision. In any event some economists argue that market prices do not adequately reflect consumers' willingness to pay.

A considerable number of empirical studies have considered fishing and thus economists have made some contribution to assisting planners and managers in making decisions on fishery provision, policy and management. However, this contribution should not be overestimated, it has been on a theoretical plane rather than a practical one, and the debate on the efficacy of the methods advocated continues. A further difficulty is that there is no clear agreement among economists on what constitutes the recreational value of resources, so that, depending on researchers' view, different methods have been considered appropriate.

Moreover, there are considerable differences between circumstances in the U.S. and Europe and consequently the emphasis placed on particular methods. In the U.S. the importance of recreational marine fishing is much greater, one result of which is that there is more conflict between it and commercial interests. Also, fishery acts set down very clearly the objectives of fishery stocks management. Therefore, U.S. research has devoted much attention to the optimal use of resources and the methods advocated reflect this. Another difference in fisheries in the two continents is the greater preponderance of privately owned resources in Europe so that the concern over the allocation and best use of resources has not been seen as either necessary or urgent.

This paper reviews a number of studies of fishery resources, with particular reference to those presented at this symposium, concentrating on the methods proposed by which benefits and costs may be measured and indicating the more useful lines of development currently being pursued. It is not concerned with fishery studies which act as a vehicle for the identification and formalization of recreation economics concepts and problems.

METHODS OF ASSESSING RECREATIONAL DEMAND AND BENEFITS AND THEIR RELEVANCE TO FISHING

The categorization of methods is not easy. Here it is based on previous reviews and a broad division of those which are macro or aggregative on the one hand, and micro or specific on the other. The distinction is not always clear so that a degree of arbitrariness creeps in. For example, systems analysis has been used at almost a national level in some research programs and as gravity models for very local studies in others. Another problem is that economists may use methods in combination; Talhelm (1972) has constructed a model using elements of the travel cost, household production function and systems analysis approaches.

The criterion adopted for considering one method as being distinct from another is essentially that researchers refer to it as such in making reference to previous work in their publications. Methods which do not fit comfortably into either a macro or micro category are listed separately.

Aggregative studies which have been on a national, regional or subregional basis have tended to be developed by geographers and planners and indicate broad trends in the level and pattern of recreational activity. At the other end of the scale are local, site/facility or activity specific studies which have been of much more interest to the economist, attempting to estimate the benefits of recreation and evaluate the resources so used.

Aggregative studies:

1. time budget analysis;

2. participation and recreation type surveys;

3. socio-economic profile studies;

4. systems analysis;

5. planning standards approach.

Local, site/facility or activity specific studies:

1. gravity models/intervening opportunities models/travel cost (Clawson) method;

2. gross expenditure estimation;

3. household production function approach;

4. direct evaluation or interview method;

5. market valuation method;

6. value of product method;

7. cost of supply estimation;

8. land values models.

Other methods:

1. Delphi technique;

2. recreation multiplier models.

Aggregative Studies

The main purposes of aggregative studies have been to identify the more important variables determining demand and to forecast its level and pattern. Very often information on time budgets participation and socio-economic profiles is collected in a single home based study. Relatively early examples of these kings of approaches are the Outdoor Recreation Resources Review Commission (ORRRC) (1962) studies in the United States and the Pilot National Recreation Surveys in the U.K. (British Travel Association 1967, 1969).

Time budget studies are an attempt to establish the likely time available for active outdoor recreation, more or less as a residual after committed time for work, shopping, domestic activities, etc., has been established. Participation surveys relate to peoples' area of residence and, for a given population, indicate the level and pattern of involvement in recreational activities. In general, a distinction is made between passive and active forms of recreation, but more sophisticated studies have examined the extent to which activities can be grouped by the predisposition of participants to undertake them. Such an exercise also relates groups of activities to the land and spatial resources likely to be used. Active pursuits will tend to require specific facilities and thus will be ‘resource based’ while more informal and passive activities will be less resource specific, i.e., demand will be what is known as “user orientated”. Apart from fishing, anglers also tend to participate in field sports, camping, nature study and sailing. Coppock and Duffield (1975) produce some evidence of such relationships in a study of recreation in the U.K.

Socio-economic characteristics are closely linked with the idea of participation in certain activities. There is much evidence from studies in both the U.S. and Europe to support this. In general, participation in active pursuits is influenced by socio-economic characteristics, whereas for passive recreation the influence is less marked. There is also some difference in the effect of such variables on urban as opposed to rural recreation; the influence is much stronger in the latter case. Thus a participant in active rural recreation is likely to be middle aged, in a professional occupation of high social status and income, and of high educational attainment.

Socio-economic profile studies have often included the collection of data on recreational expenditure.

Planning standards approaches are essentially the calculation of minimum requirements for sport facilities and recreational open space, based on the population size and structure of a given area. They are very loosely related to participation studies in that some attention is paid to the level and pattern of sport and recreational activities. They are largely concerned with the provision of facilities in urban areas.

Systems analysis is more of a technique than a method and has been advocated by geographers on the grounds that outdoor recreation is fundamentally a spatial interaction between demand and supply. The analysis concentrates on the recreational journeys, the structure of which is that demand generates travel to recreational sites (supply) giving rise to a distribution of journeys. Systems approaches can be applied on a wide-ranging scale. At a regional or sub-regional level they attempt to solve the problem of the relative attractiveness and interdependence of sites, and differences in the type of journey generated from concentrations of population. At a very local level, particularly in the elementary gravity model form, they are almost equivalent to the Clawson travel cost method for evaluating demand to specific sites.

The value of participation studies, though they are largely a description of the level and pattern of past behavior, can offer guidance on the growth or decline of demand for particular activities in the future which can assist the planning and development of facilities. They also show that demand for recreation is relatively stable and principally determined by economic and social factors.

Participation studies in Europe indicate that with the exception of informal recreation, such as walking, fishing is one of the most popular outdoor active pursuits. Specific fishing studies at a national level have built up a profile of the angler, especially with regard to the type of fishing, number of day and holiday trips and expenditure on equipment, travel and food, for example: the U.S. National Survey of Fishing, Hunting and Wildlife Associated Recreation, conducted approximately every 5 years, the National Angling Survey in the U.K., originally undertaken in 1971 and currently being repeated, and the survey of sport fishing in The Netherlands. In the U.S. special surveys have been used for specific areas or types of fishing.

In systems approaches, the emphasis to data has been on informal recreation journeys and only insofar as anglers need to travel to undertake their chosen activity are they relevant to the assessment of demand for fishing. Models specifically concerned with fishing trips are now being applied in the U.S. and Canada, but so far little development of this approach has occurred in Europe.

In general, given the present ‘state of the art’ of the economic analysis of recreation, it must be concluded that aggregative models are only marginally applicable to fishing demand and benefit estimation. They can give an overall picture of trends but cannot indicate the pattern of demand, the best locations for siting fishery resources or evaluate them. Furthermore, though the position is improving with the move back to aggregative methods, at present there is a lack of data which would facilitate the development of models. For example, in many countries there is little information on the origins and destinations of anglers by which systems approaches could be applied.

Moreover, the many different types of fishing ought to be considered separately. Sea, game and coarse fishing are almost different sports, notwithstanding the predelication for some anglers to pursue all three. Coarse fishing may also be classified by the type of resource at which it is undertaken; fishing a reservoir is very different from fishing a river, canal or water-filled quarry. There are seldom adequate data on the types of fishing and resources and, what is more disturbing is that the need to make the distinction in collecting information is often not recognized.

The U.S. excepted, aggregative approaches have also tended to underestimate the importance of obtaining information on anglers' knowledge of facilities, motivation and social attitudes to recreation, and certainly little attempt has been made, on a national scale, to collect data on the benefits participants see themselves as deriving from their chosen activity. Yet another failing is the tendency to ignore supply factors. Ostensibly fishing is an activity for which demand generation is sensitive to supply changes, as has been demonstrated in both the U.S. and Europe in connection with the creation of water supply reservoirs. Moreover, until recently little attempt has been made to ascertain anglers' attitudes to substitution between both fishing sites and types of fishing, particularly where fishery resources are lost or under threat of extinction.

A possible fruitful line of connection between aggregative and site specific studies may well be at a regional or sub-regional level where information on participation can be collected with site demand studies within the same area. Such an attempt was made in the Study of Informal Recreation in South East England (Countryside Commission 1977).

Stabler and Ash (1977) found for fishing that the national studies in the U.K., which contained data on expenditure, were useful in estimating the total benefits enjoyed by anglers for specific sites.

The value of aggregative studies is greater to the planner and manager, concerned for example with the problems of accommodating the number of anglers, optimising yields, or controlling fish-take to preserve stocks since data on size and type of catch and numbers of anglers can indicate the level of demand and stock losses, etc., then to the economist who is interested in benefit evaluation. Thus as an aid to management the development of aggregative approaches to data collection should be welcomed. The economist needs to make his requirements known before such methods can also assist him.

Local, Site/Facility or Activity Specific Studies

The methods for evaluating the benefits of specific sites or facilities are essentially set in a cost-benefit framework. Accordingly economists argue that all benefits and costs should be identified and evaluated. A number distinguish between primary and secondary benefits and costs.

Primary benefits and costs are those which apply to a specific site or facility, or accrue directly to its users. Secondary benefits and costs are indirect and can be both site or user specific but may also be nonspecific, i.e., they are externalities. (An externality is an uncompensated loss or unpaid gain.) Examples of secondary benefits are: the pleasure derived by one class of site users from the presence of another class, say, the interest generated for informal visitors by boaters on a navigable waterway; the enjoyment by anglers of fishing in pleasant surroundings; the improvement in health of participating in active recreation; the reduction in crime; the generation of income and employment in a locality as a result of recreation provisions; the increase in land values surrounding recreational areas. Secondary costs are less often identified but are nevertheless equally as relevant, for example: ecological deterioration of sites or areas; chemical pollution; noise; congestion.

The unpriced or nomimally priced nature of much recreation makes it very difficult to quantify primary benefits and costs let alone express them in monetary terms. These difficulties are compounded for those which are secondary so that economists have tended to ignore them.

The concern over primary and secondary benefits reflects the tendency for many economists' wish to evaluate total demand or supply, i.e., an all or nothing approach, whereas some (e.g., Knetsch and Davis 1974) would emphasise the importance of incremental or marginal changes in costs and benefits in making decisions about recreational resources. The author and others would take issue with this view particularly where resources are at risk; this matter is discussed below.

All the approaches listed in one way or another attempt to identify and evaluate at least the primary benefits. Except for (d), indirect means of making an evaluation have been used. A number of approaches, (e), (f), (g) and (h), can be dismissed as either theoretically suspect or impracticable, while others, such as (c), have not been developed sufficiently to be operational.

The market valuation method measures the charges judged to be the value of the recreation services produced which, when related to the number of users, yields a total value of the resource. Two problems arise, however. First, few recreational facilities are provided through the market, and, second, as indicated in the introduction, market prices do not necessarily represent the willingness to pay for recreation.

The value of the product method assumes that some tangible and monetary quantifiable output results from the recreational activity. Thus the recreational benefits of fishing would be the value of the fish caught. Clearly, this implies that anglers' only objective is to catch fish; there is much evidence that this is not so.

The cost of supply method is theoretically unacceptable because it leads to the illogical conclusion that the greater the expenditure by the supplier the greater the benefit (This is not the case where the consumer spends more since in this case expenditure represents a willingness to pay.) The land values models and recreation multiplier models respectively, are theoretically acceptable as indicated above since they represent the benefits associated with recreation. It is clear, however, that each is likely to rise to an underestimation of the benefits for they consider only one of the many benefits which might be derived from a specific facility or activity. Another, and perhaps more serious, problem is that neither method is really operational because it is not possible to identify the extent to which changes in land values or income and employment are related to specific facilities, sites or activities. Furthermore, it is not possible to establish how far increases in values or the generation of income and employment for a particular activity or location are merely transfers.

Thus, four methods—gravity/travel cost models, expenditure estimation, the construction of household production functions and direct valuation method—have been considered seriously by economists. Of these the travel cost approach in its various forms has dominated benefit estimation research, at least until recently. The method first suggested by Clawson (1959) was originally considered by economists to be basically sound conceptually and operationally but of late it has been heavily criticized.

The travel cost method will not be examined in detail here. The reader is referred to the preceding paper by the author submitted as an experience paper to this symposium for a discussion of the travel cost method and related approaches. There are also recent reviews which may usefully be consulted, Gibson (1978), Langford and Cocheba (1978) and Baxter (1979). However, the major complications of recreational activity, which the method fails to accommodate adequately, can be summarised here and related to the problems of evaluating fishery benefits. These are:

1. the interrelationship of sites,

2. the effect on demand for and evaluation of a given site of the provision or loss of a new site or sites,

3. the effect of tourism in a locality on demand for facilities,

4. the effect of pricing policies on demand, particularly dual pricing schemes—season tickets and day tickets,

5. the contrast between urban and rural recreation patterns,

6. the length and utility of the journey,

7. the assessment of site capacities and effect of congestion,

8. the assessment of the effects of quality change, other than congestion, such as the number and variety of fish present, cleanliness, natural beauty and facilities available,

9. the degree of substitutability between sites and activities.

The criticisms which economists have made of the travel cost method, and desire to incorporate the complications given above, apply equally to other methods but to an extent reflect an obsession partly with theoretical minutiae and partly the goal of constructing an all embracing general model. It is becoming increasingly apparent that recreational patterns are far too complex to be forced into a single model. On the other hand, to be operational, models need to be simpler, not more sophisticated. Moreover, the method of analysis needs to be related to the purpose of the study. Gibson (1978) argues that when the main concern is to evaluate whether one facility is worth preserving or past investment needs to be justified, knowledge of the demand curve for that facility is sufficient, whereas if it is consideration of whether a new facility should be introduced, then a number of demand curves need to be estimated. The implications of Gibson's observations are that not only should a distinction be made between total evaluations and incremental studies, but also that, to date, the travel cost approach has proved inadequate for either purpose.

Thus, though a number of economists continue to attempt to refine the travel cost approach, it has been found wanting on three counts. First, it is theoretically unacceptable because it does not fully accommodate the complexity of the recreational experience; second, it is empirically unworkable and, third, as recreation resources have become more scarce, and therefore more valuable, it has failed to capture the divergence between willingness to pay and society's evaluation of those resources.

Researchers attempting to evaluate benefits have recognised fishing as a sport which exemplifies both the complexity of recreational activity and one where its resources are very much at risk. As such they have begun to devise new methods of assessment.

Talhelm (1972), quite early in the development of recreation economics, argued that the price of fishing, from the angler's point of view, included not only the monetary cost but also the opportunity cost of time incurred for the purpose of angling. He also saw the angler as the producer as well as the consumer (in effect incorporating in his model a household production function approach) and he defines supply as the costs anglers must meet to fish at a site. Talhelm's approach had enabled him to consider the effect of supply and quality changes because estimated demand and supply functions are set in the context of the resources available in an area (defined by angler's willingness to travel to fish specific sites). Furthermore, Talhelm was critical of the traditional economics approach, considering it too concerned with demand and also indicating confusion in the distinction between the supply of fishing and the supply of fishing sites. Talhelm's approach, which is probably the most comprehensive model constructed to date, is currently being applied to recreational fishing in the South Wales area of the U.K. by Mawle and Randerson (1980). However, differences in the price structure in the U.S. and the U.K. have necessitated modifications to Talhelm's approach.

Stabler and Ash (1977) have found, for canal fishing in the U.K., that the travel cost approach was inappropriate because of the patterns of demand and nature of the resource. They adopted an approach very similar to Talhelm's by attempting to establish the price of canal fishing from outlay on tackle, boat licences and travel as well as travel and on-site time. Unfortunately, because of data problems they have not been able to develop their models to assign values to canal fishing resources as opposed to benefits derived by anglers from their fishing experience. Moreover, their work has been misinterpreted by economists who have purported that they used a gross expenditure method, which has been rejected as being conceptually unsound. However, from a management viewpoint expenditure or outlay approaches have merit in that they are relatively simple, and if other recreational opportunities are evaluated on a similar basis, allow comparisons of relative value to be made.

The household production function approach derives from the work of Becker (1965) and fishery resource management has been extensively studied within its context during the late 70s—see, for example, Anderson (1982), Bockstael and McConnell (1980), and Sutinen (1980).

The approach views consumers as producers in that goods and services are purchased as inputs into self production function because of their utility bearing attributes or characteristics. In effect, consumers incur marginal costs, certainly on an opportunity cost basis, in selecting a particular array of goods and services. Households therefore seek to maximise utility within their income constraint, or minimise costs for a given level of utility. One attraction of the method is that it easily accommodates the notion of the cost of time, which is particularly important in the analysis of recreational activities.

The method is particularly relevant to fishery resources where both recreational and commercial exploitation occurs and a decision has to be taken on the optimal use of such resources. The recreational angler, when the added benefits of additional fishing are not sufficient to cover the foregone satisfication of other uses of money and time, will no longer fish. Likewise the commercial angler would normally cease operations when marginal cost exceeds marginal revenue.

What concerns both managers and researchers is that unregulated access leads to non-optimal use of fisheries. Increased effort may increase current landings but imposes future costs by lowering fish stocks and ultimate yields and benefits, which individual anglers do not perceive. The prescription is that reducing current use will increase future benefits.

Optimal use of a resource will be achieved when the sum of net benefits for both recreational and commercial anglers are maximized over all time periods, and when their respective marginal net benefits are equal.

The problem with the household production function approach lies in estimating the marginal benefit and cost relationships of recreational and commercial anglers. Modelling has got little beyond the stage of formalizing the concepts and development of models depends on assumptions which may not hold in an empirical context. Moreover, there are complexities in the real world, particularly with respect to policy decisions which may preclude the observation of benefits and costs. Bockstael and McConnell (1980) recognize the problems associated with the approach and outline possible avenues for further research.

Theoretically, at least, the household production function approach has merit in considering the importance of the opportunity cost of income and time of undertaking recreational activities. However, it is somewhat difficult for the European researcher and manager to see its relevance because of its close association in U.S. with joint use of resources which is of much greater concern in that country. A separation of the method from this association and empirical application and testing is required before it can be fully assessed.

Because of the problems with more traditional economic approaches, increasing interest has been shown in the U.S., and to a lesser extent in the U.K., in direct methods of evaluating recreational benefits. In essence the method is to question recreationalists in an interview, on their evaluation of the resources they use.

Traditional economists have criticised it for being too hypothetical and subject to bias (e.g., P.A. Samuelson 1954). It is argued that respondents will tend to underestimate willingness to pay if they think they will escape any charges which might be introduced. Conversely, they will overestimate willingness to pay if no charge is likely to be made and they wish to see the resource preserved.

However, in an early study of the demand for forest recreation, Davis (1963) showed that used in conjunction with other methods, as external validity checks, that the approach did not result in indue bias. Bohm (1972) largely confirms this finding and since then a number of other recreational studies of this nature, including some on fishing, have been conducted (see Meyer 1979).

In the U.K., Stabler and Ash (1978) used a similar approach in their study of informal activities on canals.

The direct approach has a number of virtues. First, it is adaptable in that respondents can be asked about their evaluation of different types of fishing and resources. For example, they can be asked about their use of publicly and privately owned resources (which are likely also to have widely varying price structures), sea, game and coarse fishing, and their use of different locations. Second, interviews can be conducted simultaneoulsy with aggregative data surveys. Third, several measures of evaluation can be utilised in the same interview as validity checks. Fourth, in the face of increasing pressure on and loss of recreation resources, as is indeed the case in sports fishing, the method is likely to be more sensitive in reflecting upward revaluations by anglers of fishery resources over time.

Several economists, notably Gordon and Knetsch (1979) and Meyer (1979) in what amount to review articles, have shown that economic notions of willingness to pay and willingness to sell (i.e., the compensation required to forgo use of recreation resources) as being virtually identical, do not hold. In a number of studies willingness to sell valuations have been 3–20 times greater than those for willingness to pay. In the past economists have considered such valuations as irrational and therefore have tended to discard them from their data set. It is increasingly argued, however, that such responses are not necessarily extreme and therefore should be retained.

In addition to turning to direct approaches for estimating benefits, economists are considering multiplier models, hitherto used only in regional economics and studies of tourism to establish the income and employment generated by expenditure in designated areas. The approach is beset with empirical problems of which the most intractible are those concerned with identifying the benefits associated with particular activities or resources and the specification of the area so influenced. At present no reliable estimations of fishing resources using this method have been made.

Methods from other disciplines are also being applied to recreation such as the Delphi technique, whereby panels of experts reach a consensus on recreation values through the dissemination of continually revised estimations by individual members of the panel. Zuboy (1982) has used this technique in estimating fish yields.

The outline of the methods used by economists given above is not exhaustive for attention has been concentrated on more recent developments which are likely to prove most useful.

SUMMARY AND CONCLUSIONS

Though much progress has been made in developing methods for evaluating the benefits of recreation, economists have not yet reached a point where they can use them to make the contribution that they should to the planning and management or recreation resources. The reasons for not having achieved the goal, which seemed easy to attain in the early years of recreation economics, stem partly from economists themselves and partly from other factors.

It is unfortunate (the author is writing as one himself) that economists have displayed an element of arrogance in thinking they alone held the key to solving resource evaluation problems with little or no appreciation of planning and management needs and virtually no consultation. They have held on too long to traditional approaches, vainly pursuing the goal of a general theory of evaluation. Moreover, they have failed to recognise the complexity of many recreational activities and in many cases have obfuscated issues in their confusion and disagreement over defining recreational value. Furthermore, researchers on each side of the Atlantic have worked very much in isolation. It is seldom that European work is quoted in the U.S. literature, while certain methods well known in the U.S. have not been applied in Europe. For example, to data there is only a single case of the use of Talhelm's method and none of the application of the household production function approach in the U.K.

In their defense, it must be emphasised that economists have been researching recreation for a relatively short time and certainly in many European countries where it is given a low priority, few are working on it. (In this connection an argument put forward by A. Tuomi earlier in this symposium is appropriate; he asserted that the allocation of resources to recreation will never be given serious consideration until such time as it appears in national accounts.) Usually economists have had to use rather poor secondary data which has hampered the development of models. Furthemore, notwithstanding their shortcomings, recreation economists have been more successful in developing methods and making estimates of benefits and costs than colleagues working in such fields as environmental quality, noise and chemical pollution.

For their part planners and managers have often had too high an expectation of what economists could achieve, or the findings were not in accord with their perception of the circumstances. They also ought to accept some responsibility for the slow development of methods. Sometimes they are unable to define management objectives, do not see the need to keep records or monitor the use of their resources, and are unwilling to experiment with pricing policies and resource management techniques, to allow the effect of such changes to be observed and analysed.

It is evident, however, that the position is improving. Governments and agencies are increasingly recognizing the importance of recreation and the need for research into it to be conducted. Recent studies indicate that greater progress in estimating recreation benefits is being achieved. Of the approaches now being adopted, those which set activities with a comprehensive framework, such as Talhelm's, or concentrate directly on users' own evaluation hold the most promise, also suggesting that economists are more aware of the need to match the method to the type of activity and resource, the purpose of the study and practitioner's needs. It is too early to judge whether other methods such as systems analysis, the household production function approach, and Delphi technique will prove equally fruitful.

Somewhat timorously, the following recommendations are offered:

1. There should be more communication between economists working in the U.S. and Europe. Where the resources are similar and there are common problems it would be rewarding for researchers to cooperate in devising, developing and applying methods for evaluating recreational benefits. Certainly the exchange of ideas through appropriate publications or the exchange of personnel from time to time would be desirable.

2. More data should be collected both on an aggregative and specific resource basis. The kind of data of most value to economists would be on socio-economic characteristics; fishing experience—frequency, duration, type, locations, outlay or tackle licences, travel, etc.; anglers' own evaluation of their sport; anglers' perception of the cost and quality of fishing; catch; use of site—numbers, duration of stay; distribution of users, etc. Managers could initiate the collection of at least some such information through normal day-to-day running of the resource and consider the possibility of periodic samples of anglers and monitoring of site use. With the greater complexity and interdependence of human and eco-systems there is a need for more information so as to reduce the degree of uncertainty in making decisions on resource allocation and use.

3. A complementary approach should be adopted—the use of several methods simultaneously as validity checks. There is no reason why, say, the travel cost and related approaches should not be used in conjunction with direct methods of evaluation and the Delphi technique. It is also important that more holistic methods such as systems analysis should be combined with site specific methods; and that the methods should be related to the type of activity. For example, fishing at a large reservoir in a remote rural area may well be analyzed by a travel cost mode, whereas in an area where many fishery resources are present a systems analysis may be more appropriate. Bodies responsible for the management of resources should encourage the development of methods by their willingness to fund further research.

4. There should be a multidisciplinary approach to recreational research and resource management. The economist required to collect his own data is hardly competent to devise and execute a survey where recreationalists are asked about motivations, attitudes and perceptions; the social psychologist, sociologist and statistician should be involved. Equally, the natural scientist is not really equipped to analyse a problem from an economic viewpoint, nor the manager to set objectives with an inadequate grasp of say ecological factors.

A final comment worth making is that however much administrators, planners and managers may view economic estimates of recreation value as inadequate, in the final analysis they recognise that no value can be assigned without quantification and that ideally this quantification should be in monetary form. Thus the economist has a fundamental contribution to make to the planning and management of fishery resources.

LITERATURE CITED

Anderson, L.G. 1982 An economic analysis of joint recreational and commercial fisheries. (This Symposium.)

Baxter, M.J. 1979 Measuring the benefits of recreational site provision: a review of techniques related to the Clawson method. London, Sports Council. 47p.

Becker, G.S. 1965 A theory of the allocation of time. Economic Journal, 75:493–517.

Bockstael, N.E., and K.E. McConnell. 1980 Estimating and using the household production function for wildlife recreation. (This Symposium.)

Bohm, P. 1972 Estimating demand for public goods: an experiment. European Economic Review, 3:111–130.

British Travel Association—University of Keele. 1967, 1969 Pilot national recreational survey. Reports No. 1 and No. 2. London, University of Keele.

Clawson, M. 1959 Methods of measuring the demand for and value of outdoor recreation. Washington, D.C., Resources for the Future. Reprint No. 10. 36p.

Coppock, J.T., and B.S. Duffield. 1975 Recreation in the countryside. London, Macmillan, 261p.

Countryside Commission. 1977 Study of informal recreation in south east England: summary report for the site studies. Cheltenham, Countryside Commission. 31p.

Davis, R.K. 1963 The value of outdoor recreation: an economic study of the Maine wood. Ph.D. dissertation. Harvard University, 171p.

Gibson, J.G. 1978 Recreation land use, pages 68–96 in D.W. Pearce, ed. The valuation of social cost. London, Allen and Unwin.

Gordon, I.M. and J.L. Knetsch. 1979 Consumers surplus measures and the evaluation of resources. Land Economics, 55:1–10.

Knetsch, J.L., and R.K. Davis. 1974 Comparison of methods for recreation evaluation, pages 151–166 in Van Doren et al., eds. Land and leisure. Chicago, Maaroufa.

Langford, W.A., and D. Cocheba. 1978 The wildlife valuation problem: a critical review of economic approaches. Canadian Wildlife Service occasional paper No. 37. 52p.

Mawle, G.W., and P.F. Randerson. 1980 A method for the assessment of demand for recreational fishing in the South Wales area of the U.K. Cardiff, Department of Applied Biology. UWIST. University of Wales. 25p.

Meyer, P.A. 1979 Publicly vested values for fish and wildlife: criteria in economic welfare and interface with the law. Land Economics, 55:223–235.

National Angling Society. 1971 London, NOP Market Research Ltd. 124p.

Outdoor Recreation Resources Review Commission. 1962 Participation in outdoor recreation: factors affecting demand among American adults. Washington, D.C., O.R.R.R.C.

Samuelson, P.A. 1954 The pure theory of public expenditure, Review of Economic Statistics, 36:387-9

Stabler, M.J., and S.E. Ash. 1977 The amenity demand for inland waterways: angling. Amenity Waterways Study Unit, Reading, University of Reading. 115p.

Stabler, M.J., and S.E. Ash. 1978 The amenity demand for inland waterways: informal activities. Amenity Waterways Study Unit, Reading, University of Reading, 82 p.

Sutinen, J.G. 1980 Economic principles of allocation in recreational and commercial fisheries. (This Symposium.)

Talhelm, D.R. 1972 Analytical economics of outdoor recreation: a case study of the southern Appalachian trout fishery. Ph.D. thesis, Raleigh, North Carolina State University, 202p.

Tuomi, A.L.W. 1980 Integration of systems: discussion note on methods. (This Symposium, Panel III.)

Zuboy, J.R. 1982 The Delphi technique: a potential methodology for evaluating recreational fisheries. (This Symposium.)

EFFECTS ON FISHERIES OF ABSTRACTIONS AND PERTURBATIONS IN STREAMFLOW

Clair B. Stalnaker

Cooperative Instream Flow Service Group, U.S. Fish and Wildlife Service, 2625 Redwing Road, Fort Collins, Colorado 80526 USA

ABSTRACT

Effects of stream flow perturbations are summarized from a fisheries viewpoint and the pertinent recent literature identified which documents such effects. Methods used for evaluating instream flow requirements for fisheries are reviewed and presented in three categories: 1) those suitable for early planning and general guidance, usually based upon hydrologic data; 2) those used in water allocation processes and flow regulation schemes, usually based upon stream channel by hydraulic measurements; and 3) those suitable for detailed ecological evaluations of impacts, usually based upon regression analyses of stream attributes vs. fish population attributes. Finally the stream microhabitat as niche space is discussed along with the need to identify three separate publics in presenting fishery-flow information. These are the aquatic ecologist, the hydrologist-engineers and the administrators who act on behalf of the public.

RÉSUMÉ

Les effets des perturbations du débit des rivières sont résumés d'un point de vue piscicole, et la littérature pertinente récente qui documente de tels effets est identifiée. Les méthodes utilitées pour l'évaluation des besoins débimétriques des pêches sont examinées et présentées en trois catégories: 1) celles convenant à la première plannification et comme guide général, généralement basées sur des données hydrologiques; 2) celles utilitées dans les processus d'allocation de l'eau et dans les plans de régulation des débits, basées généralement sur des mesures hydrauliques sur un chenal de rivière; et 3) celles convenant à des évaluations écologiques détaillées des impacts, basées généralement sur des analyses de regression des caractéristiques des rivières vs. caractéristiques des populations de poissons. Finalement le microhabitat d'une rivière en temps qu'espace de niches écologiques est discuté de même que le besoin d'identifier trois différents publics lors de la présentation de l'information concernant la pêche et les débits. Ceux-ci sont les écologistes du milieu aquatique, les ingénieurs-hydrologistes et les administrateurs qui agissent au nom du public.

INTRODUCTION

Evidenced by the organization and conduct of this international symposium on fishery resource allocation problems and the considerable worldwide literature appearing over the past three decades, it is clear that serious competition and conflict for uses of freshwater exist. It is also evident that society places a high value upon stream fisheries both from a recreational as well as a commercial use. Because of the high value placed upon stream fisheries and in particular upon the salmonids (trout and salmon), much emphasis is presently being placed upon constraining water and land development schemes in some manner to preserve a portion of flowing water in these valued streams to maintain the fishery at some healthy level of production.

Traditionally, fishery management organizations have been concerned with managing or controlling water pollutants as well as the catch and recruitment of the fish populations. Due to the increased technological capability of man, it is now possible to control and regulate the world's surface water flow to such a degree that the fishery management organizations must now become involved with the water allocation and regulation decision processes of various governments. In fact, inland fishery management must now include water management as a routine endeavor. I will attempt in this paper to discuss the fishery management aspects of water allocation and use emphasizing the effects of development schemes for hydro-electric production and irrigation upon anadromous and resident salmonid fisheries.

The usual progression of impacts imposed by man upon river ecosystems has been in three phases. First, pollution brought about by municipal and industrial uses of water in or near urban population centers. Such consequences are the topics of other papers of this symposium. Second, the construction of wiers and locks accompanied by channelization and dredging for navigation on the major river systems for the purpose of transporting goods between population centers. Third, control and regulation of river flows for energy production, protection from floods, and abstraction for irrigating agricultural lands. On every continent of the earth (except Antarctica) the progression has proceeded well into the third phase.

Stanford and Ward (1979) in the proceedings of the first international symposium on regulated streams stated, “that stream regulation has exerted more profound and irrevocable effects on the character of the world's rivers than pollutants. Altered ecosystems below dams and diversions are now the most prevalent lotic environments on the earth.”

Adverse impacts resulting from the first two phases are much more ameniable to restoration because they are observed first-hand by a majority of the public (fish kills, silt deposits, sloughing banks, etc.) and engineering solutions can be brought to bear without altering the allocation of uses (pollution treatment, fish ladders, etc.). It is the third phase of water resource development that has the greatest potential for irreversible damage to the river ecosystems and the fisheries. This is due to many reasons but the principal ones relate to the fact that much of the development for streamflow regulation incorporates massive engineering works such as large dams, complex and extensive water conveyance systems, and interbasin transfer of streamflows. Such streamflow regulation alters the pattern of land use and the structure of society's use of water within a geographical region. Also, the remoteness of much of the construction and the lack of direct visual impacts to the fisheries do not tend to produce as much emotional concern by the urban public for restoration of streamflows and lost fishery resources. This lack of public demand for such restoration is understandable when abandonment of construction works and altering of water uses and human behavior are the restoration alternatives.

The need for proper management and planning prior to such irreversible construction projects has lead to a new specialization within fishery science focusing upon instream flow management for fisheries and recreation. This requires a combined understanding of fish population dynamics, hydrology, river mechanics, and water management.

IMPACTS

Reviews pertaining to the general subject of streamflow requirements of fishes and river ecology are available. Recent books which provide excellent discussions of stream ecology have been provided by Hynes (1970) and Whitton (1975). Fraser (1972) reviewed the published literature relative to streamflow (velocity and depth) and its effects upon aquatic organisms. Shirvell (1979) reviewed the effects of flows on trout ecology and Hooper (1973) summarized methods relating to trout spawning requirements. Giger (1973) reviewed the literature relevant to requirements of salmonids (especially juveniles). Bovee (1978) reviewed the salmonid preferences for depth, velocity, substrate, and temperature and compiled life stage criteria for evaluating streamflows. Excellent discussions of river channel morphological changes associated with streamflow regulation are provided by Gregory (1977) and Dunne and Leopold (1978). Ecological aspects of large rivers and the consequences of man's activities are introduced by Oglesby et al. (1972).

The proceedings of a symposium and specialty conference on instream flow needs held in the U.S. in 1976 (Orsborn and Allman 1976) documented the impacts of flow abstractions and thoroughly discussed the technical, legal, and social problems caused by increasing competition for limited streamflow.

A detailed presentation, through case examples, of the limnology of regulated streams is found in Ward and Stanford (1979) as well as a documented discussion of the problems of stream regulation in Scandinavia, Europe, Africa, Australia, and North America. A similar review of abstraction in New Zealand is found in Scott (1979). The world's most regulated salmonid river system is perhaps the Columbia in northwestern U.S. (Stanford and Ward 1979; Raymond 1979). Several additional documented cases of the effects of flow abstractions and hydro-power development were prepared for this symposium.

The general effects of abstraction and perturbations in streamflow upon the stream fisheries are presented in Table 1 as related to high dam impoundments, direct diversions, and hydropower peaking.

Table 1. Effects of abstraction and regulation of streamflow in the instream environment.^a

^a Information is drawn primarily from Fraser (1979a), Stanford and Ward (1979), Hannan (1979), Scott (1978), Simons (1979), and Ward (1976).

METHODS FOR EVALUATING INSTREAM FLOW REQUIREMENTS FOR FISHERIES

A considerable amount of study has been devoted to analyses of streamflow as a basis of maintaining stream fishery resources. Most reported work has come from the western U.S. and has concentrated upon the salmonid fisheries. Reviews and descriptions of such methods are covered quite adequately by Stalnaker and Arnette (1976), Orsborn and Allman (1976), Ott and Tarbox (1977), and Wesche and Rechard (1980).

I will summarize the general approaches to evaluating instream flow requirements under three categories: 1) rule-of-thumb or hydrologic analyses; 2) physical habitat or hydraulic analyses; and 3) standing crop-flow analyses. Most work has been related to the second category with extrapolations carried out to develop the rules-of-thumb.

Rule-of-Thumb or Hydrologic Analyses

During the past three decades much applied research has been directed toward a better understanding of the relationships among stream-flow and stream channel characteristics (Leopold 1964 et al.). In general, the mean annual discharge is a relatively high flow in that 75% of the time the discharge in a stream is less. On the other hand, the mean annual discharge fills about 40% of the bankfull depth of the channel (Dunne and Leopold 1978).

The need for rule-of-thumb procedures has come from the water planning and water administration professions for which there is usually a good historical data base of streamflow and watershed (catchment basin) runoff records.

Such methods have given rise to the “minimum” flow concept for allocating water resources among offstream and instream uses and are useful when evaluating water availability on an annual basis for planning purposes and granting of water rights for abstraction under legislative processes. Most fishery administrators do not approve of the use of rule-of-thumb derived “minimum” flows for fishery maintenance when a stream is regulated (controlled by dams or diversion) (Howie 1979; Fraser 1979b; Miller 1976; Stalnaker 1980). Tessman (1980) stated, “the best minimum flow model is one that mimics nature… The year is a continuum of cyclic events to which natural community is adapted. Minimum flow expressed as total volume of instream requirements during the course of a year is meaningless unless streamflow is distributed properly during this period.”

This discontent with the rule-of-thumb approach seems to be due to the hydrologist-engineers misconception that the fishery does not require all of the streamflow during any time other than during infrequent drought conditions (Stalnaker 1979a). Such thinking has led to the unfortunate use by planners of such historic low flow values as the 7-day Q₁₀ (the lowest flow occurring for 7 consecutive days once in 10 years), the 90% exceedence flow, 10% of mean annual discharge, and even the lowest flow of record as the selected minimum flow for instream protection. Such schemes fail to recognize that the fishery is a dynamic resource which can tolerate extreme drought conditions on infrequent occasions but can not tolerate these low flows on a sustained basis without extreme reductions in the production and yield of the fishery.

For the purpose of water planning and interim determinations of the availability of water for development, the median or average monthly flow values are now accepted as more representative of the flow necessary to maintain a healthy fishery resource (Bayha 1978). In such analyses, flows in the “optimum” or “acceptable” range are much less controversial among fishery managers and ecologists. Following are examples of rule-of-thumb approaches which have been used in reconnaissance level evaluations of water resources:

Fraser (1979b): 100% of the average flow for each month of record-optimum; 79–99% of the average flow for each month of record-acceptable.

Swank and Phillips (1976): 60–100% of mean annual flow-optimum.

Tennant (1976): 60–100% of mean annual flow-optimum.

Cuinat and Demars (1980): 70–130% of the natural characteristic low flow-legal minimum

Tessman (1980): monthly minimum equals the mean monthly flow (MMF) if MMF<40% of mean annual flow (MAF). If MMF >40% MAF, then monthly minimum equals 40% MAF. If 40% MMF> 40% MAF, then monthly minimum equals 40% MAF.

Most studies indicate that flood flows are also needed to cleanse the substrate and otherwise maintain the physical integrity of the stream channel.

Physical Habitat or Hydraulic Analyses

Whenever the flow (discharge) of a stream is proposed to be substantially altered in volume or seasonal pattern, a rule-of-thumb approach is no longer appropriate and on-site studies of the fishery and the channel characteristics should be mandatory (Miller 1976; Stalnaker 1980). As many studies have documented the preference of stream fishes for particular ranges of depths, velocities, substrate size, cover objects (Bovee 1978; Giger 1973; Hooper 1973; Hynes 1970; Shirvell 1979), and temperature (Magnuson et al. 1979; Shuter et al. 1980), nearly all site specific methods proposed to date are based upon measurement and prediction of these important stream variables.

Upon close examination, all physical habitat-flow analyses can be grouped in two categories: 1) those based upon threshold conditions at critical or limiting habitat features; and 2) those based upon microhabitat features within specified (sometimes called representative) stream reaches. Most physical habitat approaches when applied to salmonid fishes focus upon passage requirements of migratory species, the life history requirements for spawning, incubation, rearing or general adult holding, or benthic macroinvertebrate production (Stalnaker and Arnette 1976). Of considerable importance is the logic by which the fishery manager determines he flow related limiting factor(s) of the fish population in the stream reach for study.

Threshold Methods

The methods require that species criteria for depth, velocity, and substrate be specified. These criteria usually take the form of a specified range. See Stalnaker and Arnette (1976) and Wesche and Rechard (1980) for summaries of reported criteria. The other necessary step is the measurement of depth, velocity, and substrate along transects placed over the stream channel. When measured at several different discharges, the ‘usable width’ across each transect can be computed and displayed as in Fig.1. Variations on this approach are described by Stalnaker and Arnette (1976) and Wesche and Rechard (1980). Another method which has often been used is the measurement of wetted perimeter at several discharges. A plot of wetted perimeter vs. discharge is produced as shown in Fig.2. Such visual methods rely upon either a peak or observed break point on the curves to establish the discharge which maximizes the “usable habitat” (i.e., the upper threshold) in the stream channel studies.

Fig. 1. Example usable width curve for a critical spawning reach (from Thompson 1974).

Fig.2. Example wetted perimeter curve for a single cross section (from Collings 1974).

Arbitrary calculations for establishing the “minimum” threshold condition have been suggested such as:1) 75–90% of the optimum value; 2) the value at which a tangent, drawn through the origin of the graph, touches the curve; and 3) the discharge which produces the maximum contiguous width along a transect having a depth ≥ some specified depth value. These “minimum” threshold values have no biological basis and are the subject of much controversy among fishery biologists. These threshold methods do not take into consideration the timing of flow in the stream channel and, therefore, should be restricted to regulated stream applications when storage in large reservoirs makes possible releases downstream for maximizing fishery habitat conditions.

Fig. 3. Example plan views for spawning area analysis at two discharges in a study reach (from Collings 1974).

Threshold methods should be restricted to planning on-site habitat analyses conducted for impact evaluations of proposed water development schemes and the development of operating rules for flow regulation.

Microhabitat Methods

These methods differ from threshold methods above in that the species criteria are weighted (Bovee 1978) and a stream reach is described in terms of the spatial distribution of the hydraulic parameters of depth and velocity over suitable substrate (Fig.3). The areal extent of suitable habitat vs. discharge is determined for several different discharges and a graph prepared as shown in Fig.4. Maximum area of suitable habitat vs. discharge is determined from these analyses but any selection of minimum levels of flow is quite arbitrary. Gore (1978) and Gore and Judy (1980) have applied microhabitat descriptive techniques to macroinvertebrate-stream-flow problems.

Recent work by the Cooperative Instream Flow Service Group has refined microhabitat analysis by developing improved hydraulic simulation models, weighted criteria for the life stages of target fish species, and the introduction of probability and stochastic process time-series streamflow data so that the habitat suitability can be displayed over time for each species-life stage (Stalnaker 1979b).

Fig.4. Example spawning area discharge curve
(from Bishop and Scott 1973).

The basic concept is that in any instant of time and small area of the stream (dA), there exists a function Ψ[H(ι)] which relates physical parameters H(ι) to the suitability of the area as physical habitat for a given species. The usability of the area is then

The term WUA is “weighted usable area” which is physical habitat index. Integrating over a specified reach of stream the weighted usable area for the reach is

The physical parameters are simulated with predictive models represented by H(ι). The variable ι includes depth, velocity, and substrate in the stream.

The equation for the function Ψ for any location in the stream is

Ψ is the habitat suitability function
V is the velocity at a point
D is the depth at the same point
S is the substrate below that same point
P_v, P_d and P_s are the functional relationships between the habitat suitability and velocity, depth, and substrate in the environment.

These functional relationships are species and life stage specific and are referred to as habitat suitability criteria (See Fig. 5).

A simple technique to solve Equations 3 and 4 is to subdivide the total stream reach area into discreet elements and assume that the functional relationship can be determined directly from fish habitat suitability criteria. This leads to the following procedure.

For discreet elements (i), the average velocity, depth, and substrate values over the element area are used to solve Equation 5. For discreet elements we have:

where v_i, d_i, and s_i are the average values for the velocity, depth, and substrate in element i. The terms K_v(i), K_d(i), and K_s(i) are the solutions for element i. Thus:

where Ψ_i is the habitat suitability function for element i.

Habitat evaluation criteria for adult brown trout
(Y axis is normalized with optimum conditions set at 1.0).

Fig. 5. Example habitat suitability criteria for brown trout (Salmo trutta) (modified from Bovee 1978).

For discharge, the velocity and depth are simulated as a function of flow. The equation for Ψ is solved for finite elements in the stream and the weighted usable area calculated using the equation

where Ψ_i is the solution to Equation 5 for the element i and A_i is the area of element i.

The elements of the stream channel are obtained by locating cross sections along a reach of river. Each cross section may have many elements which are delineated by verticals. The length of an element is one half the distance between the adjacent cross sections.

The assumption basic to the model is that a species of fish will elect to live in physical conditions that are most suitable. Although there are many important physical factors, this model includes only velocity, depth, substrate, and cover and, therefore, is applicable only to situations where these are the principal variables of concern.

Standing Crop-Flow Analyses

Attempts to develop predictive fish standing crop models appear encouraging in that regression analyses relating standing crop of fishes to habitat variables have shown that the physical parameters account for a high percentage of the observed variability.

Binns and Eiserman (1979) developed a regression model with 10 attributes (critical period streamflow; annual streamflow variation; maximum summer stream temperature; mean water velocity; cover; stream width; stream bank stability; food abundance; food diversity; and nitrates) of a stream regressed against “trout” standing crop. Biomass and habitat measurements were highly correlated (r = 0,95). Gorman and Karr (1978) developed a “habitat diversity” index comprised of three variables: depth, velocity, and substrate type. Significant correlations were found between habitat diversity and species diversity.

Nickelson and Hafele (1978) have developed regression models for coho salmon (Onchorhynchus kisutch) juveniles, cutthroat trout (Salmo clarki), and steelhead (Salmo gairdneri) juveniles. Pool volume, cover, depth, and velocity were the model variables (Nickelson and Beidler 1979). Paragamian (1980) measured density (standing crop) of smallmouth bass (Micropterus dolomieui) simultaneously with depth and substrate size at 12 different reaches in an Iowa stream. Regression of bass abundance in proportion to gravel and cobble was significant (P < 0,05). Stalnaker (1979b) reported a high correlation (r = 0,9) between biomass of adult brown trout (Salmo trutta) and calculated habitat units for stream reaches measured at low flow.

DISCUSSION

Changes which may be expected as the result of altering water flow within a stream are complex. The interrelationships among elements of the hydrologic system, though varied and complex, are relatively simple in comparison with the social, legal, economic, and institutional interdependencies. Often, the claim is made that surplus water can be removed for out-of-channel uses without dramatically affecting the aquatic organisms. This raises two important questions: How much water must remain for the maintenance of a viable fishery? How many fish will be lost if the flow is reduced by x amount? These are hard questions to answer but if information is presented in a manner which helps the decision-makers to understand losses in habitat and carrying capacity of the stream, the job is much easier. The state-of-the-art is adequate to answer the first question. The second requires some additional research specific to stream environments.

The Microhabitat as Niche Space

Hutchinson's (1978) hypervolume theory, which is defined by the n-dimensions of the environment which permits a species to exist (the fundamental niche), applies very nicely to the microhabitat requirements of fish and invertebrate species in streams. Research is resulting in an emerging understanding of the importance of the microhabitat conditions within the stream environment (Wickham 1967; Jenkins 1969; Everest and Chapman 1972).

Studies by Cummins et al. (1964), Gore (1978), Gore and Judy (1980), and Ward and Short (1978) have shown the importance of substrate size, velocity, depth, and temperature in determining the presence and density of macroinvertebrates. Alley (1977), Dettman (1978), Finger (1979), and Shirvell (1979) present excellent studies of the microhabitat preferences for fish species in streams.

When the range and optima for the parameters of velocity, depth, substrate, and cover are discovered, the physical aspects of the fundamental niche or micro-features within a stream reach may be described. Once defined, these criteria may be used with predictive models of the hydraulic parameters (Bovee and Milhous 1978) and channel form (Simons 1979) to project changes in distribution and carrying capacity of streams for fishes and invertebrates. Modeling changes in hydraulic conditions and translating such changes into microhabitat changes is now an applied science (Stalnaker 1979b; Milhous and Grenney 1980). What is needed is more research into the description of the range and optima of the niche parameters of stream species.

Temperature, chemicals, and dissolved gases constitute the most studied set of important parameters of the fundamental niche. Considerable information is available on the affects of these upon growth as well as lethal tolerances. Also, water quality modeling is quite advanced and can be used to define unfavorable conditions when streamflow is regulated. What is now needed is research into the behavioral preferences of aquatic organisms. Magnuson et al. (1979) introduced the concept of the thermal niche represented by species preferences. They further propose that the median ±33% of the laboratory determined thermal distribution will be the most useful measure of the thermal niche.

I suggest that the use of simulation modeling of water quality and physical habitat features, combined with the description of the species niche through development of range and optima criteria is necessary before the ultimate question posed by the water developer can be answered. How many fish will be lost if we reduce the flow by x amount?

A significant step forward is represented by the studies of Shuter et al. (1980) who used a stochastic temperature model and a deterministic biological model of spawning-nest survival of smallmouth bass (Micropterus dolomieui) to predict changes in year class strength over years of time. Model predictions also correlated well with the northern distribution limits of the smallmouth bass across North America.

The Presentation of Fishery Flow Information

The fishery manager must be able to communicate his findings and proposals to at least three important segments of the public:

1. The aquatic ecologists and biologists for the purpose of establishing credibility from a scientific standpoint.

2. The hydrologist-engineer to mesh the fishery requirements for water into the accepted view of the water supply and other existing uses, and proposed alterations in flow. This communication is necessary to establish that the fishery request is not “arbitrary and capricious.”

3. The general public and the decision-makers to establish the magnitude of the trade-offs associated with any proposed water management scheme. This communication is necessary to present the gains or losses of important fishery habitat in terms that relate to public use.

When applying any methodology it is important that the fishery scientist, the hydrologist, and the engineer all discuss the analysis process before, during, and after measurements are made. This assures that the appropriate information is developed in the proper format. In developing any recommendation, the stream hydrograph and the fishery/flow relationship is necessary to establish the base condition or status quo. The general public and the decision-maker need to be shown the trade-off situations in simple graphic terms. Fig. 6 shows the effects of proposed abstraction between two alternative sites for diversion while Fig. 7 shows the resource level for three alternative abstraction schemes proposed at one site.

Too often, analyses do not lay out the logic by which the recommendations and trade-offs are identified. This logical process is enhanced by a joint hydrograph-fish habitat display which considers the stochastic variability of both.

We would expect standing crop fluctuations to follow the annual variation in the amount of habitat available to each life stage. A considerable lag time often exists between the occurrence of a limiting habitat situation and the time that the limit is detected in population numbers. For example, a weak year class of fish may be the result of marginal habitat conditions for spawning 3 years earlier, or poor habitat for juveniles in one or both intervening years prior to collection.

Hydrologic process

There are two kinds of hydrologic processes: 1) probability processes; and 2) stochastic processes (Chow 1964). Displaying fishery habitat information as generated by monthly discharges in the stream reaches studied facilitates better communication between hydrologists and biologists. Probability processes are time-independent in that the concept of frequency is direct, but the monthly time-series sequences are not necessarily related one to another. For example, median monthly discharges (1 in 2 year monthly occurrence events) displayed as an annual hydrograph do not necessarily follow; i.e., the median February discharge is not likely to follow the median January discharge.

Stochastic processes are time-dependent. Both the concept of frequency occurrence intervals and a time-series relationship exist. For example, daily discharges determined from gage records may be redistributed according to some probability theory with the results averaged into weekly, monthly, seasonal, or yearly stochastic process discharges for similar habitat analyses.

Fig.6. Example of the effects of an identical abstraction of water from two streams. A is “before” and B is “after” situation.

Fig. 7. Example of the desired format for displaying the consequences of three alternative schemes (from Allred 1976).

Milhous and Grenney (1980) describe the average habitat as

where WUA is the time-average weighted usable area for a specified steam reach and specified species at a given discharge (Q) over time T.

Monthly time series analyses have been conducted in which WUA is calculated for the average flow present during each monthly period of time. This is repeated for each month of the year as well as for monthly low flow conditions with specified occurrence intervals. Fig. 8 shows such a plot for the median and one-in-ten year monthly probable hydrographs.

Annual variation in habitat can be presented as a stochastic process by computing the average index value over the appropriate season of each year. Fig. 9 presents the annual variation in usable spawning habitat over several years as a function of flow. In this case, the maximum of 3 monthly values (September, October, and November) was chosen to represent each year. Fig. 10 is an example of a habitat duration curve. From this curve we see that a river management scheme has resulted in significant reduction in the spawning habitat about 25% of the time or one year out of four. Also, the one-in-ten and one-in-twenty year events are quite drastic.

Fig. 8. Monthly time series analysis of steelhead habitat in a coastal California stream.

Fig. 9. Variation in spawning area for fall Chinook salmon in the Dechutes River, Washington (from Collings and Hill 1973).

Fig. 10. Example habitat duration curve for spawning.

Fig. 11. Recommended minimum monthly flow for the Cedar River, Washington (from Miller 1976).

Fig. 12. Monthly median and 90% exceedance flows with recommended monthly instream flow requirements (from Stalnaker 1980).

These habitat duration curves may be developed from flow records or from stochastic flow models which generate probable flow events.

Flow recommendations for fish habitat maintenance can be made which account for infrequent low flow conditions but do not force the fishery to exist under perpetual drought conditions (Stalnaker 1980). Examples of such a recommended flow regimes are given in Figs. 11 and 12.

Many of these flow-habitat displays are most useful in developing the logic behind an instream flow recommendation and are good communication devices between hydrologists and fishery managers.

The potential for irreversible damage to stream fisheries is great as evidenced by the fishery literature of the past three decades. Fishery science must take on the responsibility of concentrated research on stream microhabitat requirements on fishes and invertebrates as well as the responsibility for managing water to maximize fishery habitat within the constraints of multiple uses and conflicting demands for abstractions.

I feel strongly that rational management of the earth's flowing freshwater resources requires an interdisciplinary approach. Leonard (1972) laid it on the line when he said,

One of the most encouraging things… is recognition of the need for decision-making to become truly interdisciplinary… the specialist can no longer responsibly fall back on the Philistine attitude that he will do his thing and if they are intelligent enough to make use of his golden discoveries, well and good! but if not, it's their hard luck. The specialist—and most of us are—must continue to sharpen his individual array of tools. But he must also learn enough of the corpus of other specialties to be able to call for help when and from whatever source it is needed, and perhaps most important to all, to accord other specialties respectful consideration. What we must have is not either-or but and-and.

LITERATURE CITED

Alley, D.W. Jr. 1977 The energetic significance of microhabitat selection by fishes in a foothill Sierra stream. M. S. thesis, University of California, Davis, California.

Allred, C.S. 1976 Data needs for decision-making, pages 401–407 in J.F. Orsborn and C.H. Allman, (eds.). Instream flow needs. Volume I. Bethesda, American Fisheries Society.

Bayha, K. 1976 Instream flow methodologies for regional and national assessments. Instream flow information paper, No. 7, FWS/OBS 78/61, U.S. Fish and Wildlife Service, Office of Biological Services, Washington, D.C. 98p.

Binns, N.A. and F.M. Eiserman. 1979 Quantification of fluvial trout habitat in Wyoming. Trans.Am. Fish. Soc., 108:215–228.

Bishop R.A. and J.W. Scott. 1973 The determination and establishment of minimum stream flows in Washington State. Draft Rept. No. 73-033. Washington Dept. of Ecology, Olympia, Washington. 81 pp.

Bovee, K.D. 1978 Probability of use criteria for the family Salmonidae. Instream flow information paper No. 4, FWS/OBS 78/07, U.S. Fish and Wildlife Service, Office of Biological Services, Washington, D.C. 80 p.

Bovee, K.D. and R.T. Milhous. 1978 Hydraulic simulation in instream flow studies: theory and techniques. Instream flow information paper, No. 5, FWS/OBS 78/33, U.S. Fish and Wildlife Service, Office of Biological Services, Washington, D.C. 131 p.

Collings, M.R. 1974 Generalization of spawning and rearing discharges for several Pacific salmon species in western Washington. USGS, open-file report. 39p.

Collings, M.R. and G.W. Hill. 1973 The hydrology of ten streams in western Washington as related to several Pacific salmon species, Hymptulips, Elochoman, Green, and Wynoochee Rivers. U.S. Geological Survey. Water Resources Inv. 11–73. 149 p.

Cuinat, R. and J.J. Demars. 1980 Débit réservé et autres mesures de protection demandées lors de la création de microcentrales hydroélectriques en Auvergne et Limousin. Unpublished manuscript.

Cummins, K.W., C.A. Tryon, Jr. and R.T. Hartman (eds.). 1964 Organism-substrate relationships in streams. Special publication No. 4, Pymatuning Laboratory of Ecology, University of Pittsburgh, Pittsburgh, Pennsylvania.

Dettman, D.H. 1978 Factors influencing the distribution, abundance, and microhabitat segregation of brown trout (Salmo gairdneri) and Sacramento squawfish (Ptychocheilus grandis) in Deep Creek, California, M.S. thesis, University of California, Davis, California.

Dunn, T. and L.B. Leopold. 1978 Water in environmental planning. San Francisco, Freeman. 818 p.

Everest, F.H. and D.H. Chapman. 1972 Habitat selection and spatial interaction by juvenile chinook salmon and steelhead trout in two Idaho streams. J.Fish.Res.Board Can., 29:91–100.

Finger, T.R. 1979 Patterns of interactive segregation in three species of sculpins (Cottus) in western Oregon. Ph.D. thesis, Oregon State University, Corvallis, Oregon.

Fraser, J.C. 1972 Regulated stream discharge for fish and other aquatic resources: an annotated bibliography. Food and Agriculture Organization Fisheries Technical Paper No. 112 (FIRI/T111), Food and Agriculture Organization United Nations, Rome, Italy.

Fraser, J.C. 1979a Effects of abstraction on salmonids, pages 92–97 in D. Scott, (ed.). The effects of water abstraction on fisheries. National Water Protection Committee of Acclimatisation Society, Dunedin, New Zealand.

Fraser, J.C. 1975b Emergency response (How should administrators respond to a request for an opinion on abstraction within one month?), pages 60–66 in D. Scott, (ed.). The effects of water abstraction on fisheries. National Water Protection Committee of Acclimatisation Society, Dunedin, New Zealand.

Giger, R.D. 1973 Streamflow requirements of salmonids. Report Andromous Fish Project 14-16-0001-4150, Oregon Wildlife Commission, Corvallis, Oregon.

Gore, J.A. 1978 A technique for predicting in-stream flow requirements of benthic macroinvertebrates. Freshwater Biology, 8:141–151.

Gore, J.A. and R.D. Judy. 1980 Analysis of benthic macroinvertebrate data: predictive models of density for use in instream flow studies and regulated flow management. Unpublished manuscript.

Gorman, O.T. and J.R. Karr. 1978 Habitat structure and stream fish communities. Ecology, 593:507–515.

Gregory, K.J. (ed.). 1977 River channel changes. New York, John Wiley and Sons.

Hannan, H.H. 1979 Chemical modifications in reservoir-related streams, pages 75–94 in J.V. Ward, and J.A. Stanford (eds.). The ecology of regulated streams. New York, Plenum Press.

Hooper, D.R. 1973 Evaluation of the effects of flows on trout stream ecology. Pacific Gas and Electric Company, Emeryville, California.

Howie, W.R. 1979 Distribution of abstraction, pages 5–11 in D. Scott (ed.). The effects of water abstraction on fisheries. National Water Protection Committee of Acclimatisation Societies, Dunedin, New Zealand.

Hutchinson, G.E. 1978 An introduction to population biology. Yale University Press. 4th printing. 260 p.

Hynes, H.B.N. 1970 The ecology of running waters. Toronto, University of Toronto Press.

Jenkins, T.M. 1969 Social structure, position choice and microdistribution of two trout species (Salmo trutta and Salmo gairdneri) resident in mountain streams. Animal Behavior Monographs, 2:57–123.

Leonard, J.W. 1972 Summary, pages 455–460 in R.T. Oglesby, C.A. Carlson, and J.A. McCann (eds.). River ecology and man. New York, Academic Press.

Leopold, L.B., M.G. Wolman and J.P. Miller. 1964 Fluvial processes in geomorphology. San Francisco, Freeman.

Magnuson, J.J., L.B. Crowder and P.A. Medvick. 1979 Temperature as an ecological resource. Am. Zool., 19:331–343.

Milhous, R.T. and K.D. Bovee. 1978 The stochastic variation of instream values in rivers. Proceedings, ASCE Hydraulics Div. Conf., College Park, Maryland.

Milhous, R.T. and W.J. Grenney. 1980 The quantification and reservation of instream flows, in Preprints of the Specialized Conference on New Developments in River Basin Management, Cincinnati, Ohio.

Miller, J.W. 1976 Relating fish production to streamflow levels using fish and water management models, pages 545–563 in J.F. Orsborn and C.H. Allman (eds.). Instream flow needs. Volume II. Bethesda, American Fisheries Society.

Nickelson, T.E. and W.M. Beidler. 1979 Williamette basin streamflow studies. Oregon Department of Fish and Wildlife, Fish Research Project 2-4-8-20-02, Final job report.

Nickelson, T.E. and R.E. Hafele. 1978 Streamflow requirements of salmonids. Oregon Department of Fish and Wildlife, Federal Air Project AFS-62 Annual Progress Report.

Oglesby, R.T., C.A. Carlson and J.A. McCann (eds.). 1972 River ecology and man. New York, Academic Press.

Orsborn, J.F. and C.H. Allman (eds.). 1976 Instream flow needs. Volumes I and II. Bethesda, American Fisheries Society.

Ott, A.G. and K.E. Tarbox. 1977 “Instream flow” applicability of existing methodologies for Alaskan waters. Report prepared for Alaska Department of Fish and Game and Alaska Department of Natural Resources, Woodward-Clyde Consultants, Anchorage, Alaska.

Paragamian, V.L. 1980 Smallmouth bass—The habitat analogy. Proceedings The Warmwater Streams Symposium. Knoxville, Tennessee, Southern Division, American Fisheries Society.

Raymond, H.L. 1979 Effects of dams and impoundments on migrations of juvenile chinook salmon and steelhead from the Snake River, 1966 to 1975. Trans.Am.Fish.Soc., 108:505–529.

Scott, D. 1978 Small hydroschemes and fisheries. New Zealand Engineering, 38(8):178–179.

Scott, D. (ed.). 1979 The effects of water abstraction on fisheries. National Water Protection Committee of Acclimatisation Society, Dunedin, New Zealand.

Shirvell, C.S. 1979 The effects of abstraction on a trout stream. Ph.D. thesis, University of Otago, Dunedin, New Zealand.

Shuter, B.J., J.A. MacLean, F.E.J. Fry and H.A. Regier. 1980 Stochastic simulation of temperature effects on first-year survival of smallmouth bass. Trans.Am.Fish.Soc., 109:1–34.

Simons, D.B. 1979 Effects of stream regulation on channel morphology, pages 95–111 in J.V. Ward and J.A. Stanford (eds.). The ecology of regulated streams. New York, Plenum Press.

Stalnaker, C.B. 1979a Myths concerning instream flows: a background to understanding instream uses, pages 1–7 in Wassenberg, P.S., S. Olive, J.L. DeMott, and C.B. Stalnaker. Elements in negotiating flows associated with federal projects. Instream flow information paper No. 9, FWS/OBS 79/03, U.S. Fish and Wildlife Service, Office of Biological Services, Washington, D.C.

Stalnaker, C.B. 1979b The use of habitat structure preferenda for establishing flow regimes necessary for maintenance of fish habitat, pages 321–337 in J.V. Ward and J.A. Stanford (eds.). The ecology of regulated streams. New York, Plenum Press.

Stalnaker, C.B. 1980 Low flow as a limiting factor in warmwater streams. Proceedings Warmwater Streams Symposium. Knoxville, Tennessee, Southern Division, American Fisheries Society.

Stalnaker, C.B. and J.L. Arnette (eds.). 1976 Methodologies for the determination of stream resource flow requirements: an assessment. U.S. Fish and Wildlife Service, Office of Biological Services, Washington, D.C.

Stanford, J.A. and J.V. Ward. 1979 Stream regulation in North America, pages 215–236 in J.V. Ward and J.A. Stanford (eds.). The ecology of regulated streams. New York, Plenum Press.

Swank, G.W. and R.W. Phillips. 1976 Instream flow methodology for the forest service in the Pacific northwest region, pages 334–343 in J.F. Orsborn and C.H. Allman (eds.). Instream flow needs. Volume II. Bethesda, American Fisheries Society.

Tennant, D.L. 1976 Instream flow regimes for fish, wildlife, recreation, and related environmental resources, pages 359–373 in J.F. Orsborn and C.H. Allman (eds.). Instream flow needs. Volume II. Bethesda, American Fisheries Society.

Tessmann, S.A. 1980 Environmental assessment: reconnaissance elements of the western Dakotas region of South Dakota study report.

Thompson, K.E. 1974 Salmonids, pages 85–103 in K. Bayha (ed.). The anatomy of a river. Report Hills Canyon Task Force. Pacific N.W. River Basin Comm., Portland, Oregon.

Ward, J.V. 1976 Effects of flow patterns below large dams on stream benthos: a review, pages 235–253 in J.F. Orsborn and C.H. Allman (eds.). Instream flow needs. Volume II. Bethesda, American Fisheries Society.

Ward, J.V. and R.A. Short. 1978 Macroinvertebrate community structure of four special lotic habitats in Colorado, USA. Int.Assoc.Theor.Appl.Limnol.Proceed., 20:1382–1387.

Ward, J.V. and J.A. Stanford (eds.). 1979 The ecology of regulated streams. New York, Plenum Press.

Wesche, T.A. and P.A. Rechard. 1980 Summary of instream flow methods for fisheries and related research needs. Eisenhower Consortium Bull. No. 9. Rocky Mountain Forest and Range Experiment Station, Fort Collins, Colorado.

Whitton, B.A. (ed.). 1975 River ecology. Oxford, England, Blackwell Science Publication.

Wickham, G.M. 1967 Physical microhabitats of trout. M. S. thesis Colorado State University, Fort Collins, Colorado,.